Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A device, comprising: an optical circuit including: a plurality of light emitting diodes configured to emit light at least at a first wavelength in a range of 370 nm to 410 nm and at a second wavelength equal to or greater than 660 nm; at least one photodetector configured to detect photoplethysmography (PPG) signals in response to pulsating blood flow, wherein the PPG signals include a first spectral response obtained from light reflected at the first wavelength from skin tissue of a patient and a second spectral response obtained from light reflected at the second wavelength from the skin tissue of the patient; a signal processing circuit configured to generate PPG input data using the first spectral response at the first wavelength in a range of 370 nm to 410 nm and the second spectral response at the second wavelength equal to or greater than 660 nm; and a neural network processing device implementing a machine learning algorithm, wherein one or more parameters of the machine learning algorithm are determined using a training set, wherein the training set includes training PPG input data obtained from a healthy population and corresponding known glucose levels from the healthy population, wherein the training PPG input data includes spectral responses at the first wavelength and at the second wavelength from the healthy population and wherein the neural network processing device is configured to: determine a glucose level in blood flow of the patient from the PPG input data including the first spectral response at the first wavelength and the second spectral response at the second wavelength.
2. The device of claim 1 , wherein the second wavelength is in one of: a visible range or an infrared (IR) range.
This invention relates to an optical device designed to detect and analyze light at multiple wavelengths, addressing challenges in applications requiring precise spectral discrimination. The device includes a light source emitting light at a first wavelength and a detector configured to receive light at a second wavelength, which may be in the visible or infrared (IR) range. The device further comprises a filter system that selectively transmits or blocks light based on wavelength, ensuring accurate detection of the second wavelength while minimizing interference from other wavelengths. The filter system may include tunable elements to dynamically adjust transmission properties, enhancing flexibility in different operating conditions. The detector is optimized for sensitivity in the second wavelength range, improving signal-to-noise ratio and detection accuracy. The device may also incorporate calibration mechanisms to compensate for environmental factors or component drift, ensuring consistent performance over time. Applications include spectroscopy, environmental monitoring, medical diagnostics, and industrial process control, where precise wavelength discrimination is critical. The invention improves upon existing systems by providing a more adaptable and accurate solution for multi-wavelength optical detection.
3. The device of claim 1 , wherein the PPG input data includes: a value L λ1 generated using the first spectral response, wherein the value L λ1 isolates the first spectral response due to pulsating blood flow; and a value L λ2 generated using the second spectral response, wherein the value L λ2 isolates the second spectral response due to pulsating blood flow.
This invention relates to photoplethysmography (PPG) devices used for measuring physiological signals, particularly blood flow characteristics. The problem addressed is the need to accurately isolate and analyze specific spectral responses from PPG signals to improve the reliability of blood flow measurements. The device processes PPG input data by extracting two distinct spectral responses from the signal. The first spectral response, represented by a value Lλ1, is generated to isolate the component of the PPG signal attributable to pulsating blood flow at a specific wavelength. Similarly, the second spectral response, represented by a value Lλ2, is generated to isolate the component of the PPG signal attributable to pulsating blood flow at a different wavelength. By separating these spectral responses, the device can more accurately analyze the pulsating blood flow characteristics, reducing interference from other physiological or environmental factors. The device may include a light source emitting at least two different wavelengths, a detector to capture the reflected or transmitted light, and processing circuitry to compute the spectral response values. The isolation of these spectral components enhances the device's ability to measure blood flow dynamics, such as heart rate, oxygen saturation, or vascular health, with greater precision. This approach is particularly useful in medical monitoring, fitness tracking, and wearable health devices where accurate PPG measurements are critical.
4. The device of claim 3 , wherein the PPG input data further includes: a value R λ1,λ2 obtained from a ratio of the value L λ1 and the value L λ2 .
A device for processing photoplethysmography (PPG) signals is disclosed, addressing the challenge of accurately extracting physiological parameters from PPG data, which can be noisy and influenced by motion artifacts. The device receives PPG input data, including light intensity measurements at two different wavelengths, λ1 and λ2, denoted as Lλ1 and Lλ2. These measurements are used to compute a ratio value, Rλ1,λ2, derived from the ratio of Lλ1 to Lλ2. This ratio is a key parameter for deriving physiological information, such as oxygen saturation (SpO2), by comparing the absorption characteristics of light at different wavelengths. The device processes these values to enhance signal quality, reduce noise, and improve the accuracy of derived physiological metrics. The inclusion of the ratio Rλ1,λ2 allows for more robust and reliable calculations, particularly in scenarios where individual light intensity measurements may be affected by external factors. The system may further incorporate additional signal processing techniques, such as filtering or normalization, to refine the PPG data before computing the ratio. This approach ensures that the derived physiological parameters are more accurate and reliable for medical or fitness monitoring applications.
5. The device of claim 1 , wherein the PPG input data includes: a first AC component signal I AC generated using the first spectral response; or a second AC component signal I AC generated using the second spectral response.
The invention relates to a device for processing photoplethysmography (PPG) signals, which are used to measure physiological parameters such as heart rate or blood oxygen levels. A common challenge in PPG signal processing is accurately extracting meaningful information from raw sensor data, which can be affected by noise, motion artifacts, or varying light conditions. The device addresses this by utilizing multiple spectral responses to enhance signal quality and reliability. The device includes a PPG sensor configured to generate input data based on light reflected or transmitted through biological tissue. The input data includes an alternating current (AC) component signal, which represents the pulsatile changes in blood volume. The device processes this signal using at least two distinct spectral responses—one corresponding to a first wavelength range and another to a second wavelength range. The first AC component signal (I_AC) is derived from the first spectral response, while the second AC component signal (I_AC) is derived from the second spectral response. By analyzing these signals separately or in combination, the device improves the accuracy and robustness of physiological measurements. This approach allows for better noise rejection and compensation for environmental or physiological variations, leading to more reliable health monitoring.
6. The device of claim 1 , wherein the PPG input data includes: a plurality of systolic points and diastolic points generated using the first spectral response; and a plurality of systolic points and diastolic points generated using the second spectral response.
This invention relates to a device for processing photoplethysmographic (PPG) input data to analyze cardiovascular parameters. The device addresses the challenge of accurately measuring blood pressure and other physiological metrics by leveraging multiple spectral responses from a light source. The device captures PPG signals using at least two distinct spectral responses, such as different wavelengths of light, to enhance measurement accuracy and reliability. The PPG input data includes multiple systolic and diastolic points derived from each spectral response. By analyzing these points, the device can improve the detection of cardiovascular events, such as heart rate variability and blood pressure fluctuations. The use of multiple spectral responses helps mitigate noise and artifacts, providing more robust and precise physiological measurements. The device may be used in wearable or medical monitoring systems to provide continuous, non-invasive health tracking. The invention enhances the reliability of PPG-based diagnostics by leveraging multi-spectral analysis to extract more detailed cardiovascular information.
7. The device of claim 1 , wherein the PPG signals are generated by at least one of: a PPG circuit or a non-contact camera.
A photoplethysmography (PPG) signal acquisition device is designed to measure physiological parameters such as heart rate, blood oxygen saturation, or other cardiovascular metrics. Traditional PPG systems rely on contact-based sensors, which may be inconvenient or impractical in certain applications. This device addresses limitations by incorporating multiple signal acquisition methods. The primary PPG circuit uses optical sensors to detect blood volume changes in a user's tissue, typically via a wearable or handheld device. Additionally, the device includes a non-contact camera system that captures PPG signals remotely, such as from facial or other body regions, without requiring physical contact. The non-contact approach enables continuous monitoring in scenarios where traditional sensors are impractical, such as during sleep or in medical settings where hygiene is critical. The device may combine signals from both sources to improve accuracy or redundancy. This dual-mode acquisition system enhances versatility, allowing seamless transitions between contact and non-contact measurements based on user needs or environmental conditions. The technology is particularly useful in healthcare, fitness tracking, and remote patient monitoring.
8. The device of claim 1 , wherein the neural network processing device is further configured to: periodically receive an updated learning vector, wherein the updated learning vector is generated from an updated training set, wherein the updated training set includes updated PPG input data obtained using updated PPG signals at the first wavelength and at the second wavelength and corresponding known glucose concentration levels; and reconfigure one or more parameters of the machine learning algorithm for determining the glucose level in blood flow of the patient using the updated learning vector.
A device for non-invasively measuring blood glucose levels uses a photoplethysmography (PPG) sensor to capture signals at two distinct wavelengths. The device includes a neural network processing unit that applies a machine learning algorithm to the PPG signals to estimate glucose concentration in the blood. The neural network is initially trained on a training set containing PPG input data from the two wavelengths and corresponding known glucose concentration levels. To maintain accuracy over time, the device periodically receives an updated learning vector derived from an updated training set. This updated training set includes new PPG signals at the same two wavelengths and their corresponding glucose concentration levels. The neural network then adjusts one or more of its parameters based on the updated learning vector, allowing the algorithm to adapt to changes in the patient's physiological conditions or sensor performance. This continuous updating ensures the device's glucose measurements remain reliable and accurate. The system is designed for use in medical monitoring, particularly for patients requiring frequent glucose level assessments without invasive blood sampling.
9. The device of claim 1 , wherein the device is implemented within a user device or is configured to communicate with a user device.
A system for managing user device operations includes a processing unit and a memory storing instructions executable by the processing unit to perform functions. The system monitors device performance metrics such as battery level, processing load, and network connectivity. Based on these metrics, it dynamically adjusts device settings to optimize performance, such as reducing background processes during low battery conditions or prioritizing critical tasks under high processing load. The system may also predict future performance needs using historical data and user behavior patterns to preemptively adjust settings. Additionally, it can communicate with external devices or cloud services to offload tasks or retrieve additional processing resources when needed. The system is integrated within a user device or operates as a separate module that communicates with the user device to implement these optimizations. This approach ensures efficient resource utilization while maintaining device responsiveness and extending battery life.
10. A device, comprising: a signal processing circuit configured to: receive photoplethysmography (PPG) signals, wherein the PPG signals include a first spectral response obtained from light at a first wavelength reflected from skin tissue of a patient and a second spectral response obtained from light reflected at a second wavelength reflected from skin tissue of the patient, wherein the first wavelength is between 370 nm and 410 nm and wherein the second wavelength is equal to or greater than 660 nm; generate PPG input data using the first spectral response obtained from light reflected at the first wavelength and the second spectral response obtained from light reflected at the second wavelength; and a neural network processing device configured to: pre-configure one or more parameters using a learning vector generated from a training set, wherein the training set includes additional PPG input data obtained from a healthy population and corresponding known nitric oxide (NO) levels, wherein the additional PPG input data includes additional spectral responses at the first wavelength and at the second wavelength; and determine an NO level in blood flow from the first spectral response obtained from light reflected at the first wavelength and the second spectral response obtained from light reflected at the second wavelength.
A device for non-invasive nitric oxide (NO) level measurement in blood flow uses photoplethysmography (PPG) signals. The device addresses the challenge of accurately detecting NO levels without invasive procedures, which is critical for cardiovascular and respiratory health monitoring. The system includes a signal processing circuit that receives PPG signals containing spectral responses from two distinct wavelengths of light reflected from skin tissue. The first wavelength ranges between 370 nm and 410 nm, while the second wavelength is 660 nm or greater. These wavelengths are selected to capture specific absorption characteristics of blood components, including NO-related signals. The circuit generates PPG input data by combining the spectral responses from both wavelengths. A neural network processing device is pre-configured using a learning vector derived from a training set. This training set includes PPG input data from a healthy population, along with corresponding known NO levels, ensuring the neural network can accurately interpret the spectral responses. The neural network then processes the PPG input data to determine the NO level in the patient's blood flow. This approach leverages machine learning to enhance the accuracy and reliability of NO level detection, providing a non-invasive solution for clinical and diagnostic applications.
11. The device of claim 10 , wherein the second wavelength includes a wavelength in the IR range.
A system for optical communication or sensing uses a light source emitting light at a first wavelength and a second wavelength, where the second wavelength is in the infrared (IR) range. The system includes a modulator that adjusts the intensity or phase of the light at the second wavelength to encode data or control optical properties. The light is then transmitted through an optical fiber or free space to a receiver, which detects and processes the modulated signal. The IR wavelength may be used for long-distance communication, low-loss transmission, or compatibility with existing IR-based systems. The system may also include a controller to manage modulation parameters, ensuring efficient signal transmission and reception. The use of IR wavelengths improves signal stability and reduces interference in environments with ambient light or other optical noise. The system can be applied in telecommunications, remote sensing, or industrial monitoring, where reliable optical signal transmission is required.
12. The device of claim 10 , wherein the first wavelength is 390 nm or 395 nm.
The invention relates to a device for detecting or analyzing substances, particularly in biological or chemical samples, using optical methods. The device addresses the challenge of accurately identifying specific substances by leveraging precise wavelength selection to enhance detection sensitivity and specificity. The device includes a light source configured to emit light at a first wavelength, which is either 390 nm or 395 nm, and a second wavelength. The first wavelength is optimized for exciting or interacting with target molecules, while the second wavelength serves as a reference or control. The device further includes a detector that measures the interaction of the emitted light with the sample, such as fluorescence, absorption, or scattering, to generate a signal indicative of the presence or concentration of the target substance. The use of the specific first wavelength (390 nm or 395 nm) is chosen to match the absorption or excitation characteristics of the target molecules, improving detection accuracy. The device may also include optical components like filters, lenses, or waveguides to direct and modulate the light paths, ensuring precise measurement. The system can be integrated into analytical instruments for applications in medical diagnostics, environmental monitoring, or industrial quality control.
13. The device of claim 10 , wherein the signal processing module is configured to generate the PPG input data using the first spectral response and the second spectral response by generating characteristic features related to the shape of the PPG waveform from each of the first spectral response and the second spectral response.
This invention relates to signal processing in photoplethysmography (PPG) systems, which measure blood volume changes in tissues using light absorption. The problem addressed is improving the accuracy and reliability of PPG measurements by analyzing multiple spectral responses to extract physiological information. The device includes a signal processing module that processes light absorption data from at least two different spectral responses (e.g., different wavelengths or light sources). The module generates photoplethysmographic (PPG) input data by extracting characteristic features related to the shape of the PPG waveform from each spectral response. These features may include amplitude, timing, or morphological aspects of the waveform, which can be used to derive physiological parameters such as heart rate, oxygen saturation, or blood pressure. By analyzing multiple spectral responses, the system enhances measurement robustness and reduces noise or interference effects. The extracted features can be combined or compared to improve accuracy or detect anomalies in the PPG signal. This approach is particularly useful in wearable or medical devices where reliable PPG measurements are critical.
14. The device of claim 13 , wherein the signal processing module is configured to generate PPG input data using the first spectral response and the second spectral response by generating from each of the first spectral response and the second spectral response; one or more of: a pulse shape, average distance between pulses, variance, instant energy information, or energy variance.
A photoplethysmography (PPG) signal processing device analyzes physiological signals from multiple spectral responses to extract cardiovascular metrics. The device captures light signals at different wavelengths, generating first and second spectral responses. A signal processing module derives PPG input data from these responses by computing one or more of the following metrics for each: pulse shape, average distance between pulses, variance, instantaneous energy information, or energy variance. These metrics help assess cardiovascular parameters such as heart rate, pulse wave velocity, and vascular health. The device improves signal accuracy by leveraging multi-wavelength analysis, reducing noise and enhancing reliability in physiological monitoring. Applications include wearable health devices, medical diagnostics, and fitness tracking. The system enables non-invasive, continuous monitoring of vital signs by processing spectral data to extract meaningful physiological indicators.
15. The device of claim 10 , wherein the neural network processing device is further configured to correlate one or more parameters obtained from at least one of the first spectral response and the second spectral response to an oxygen saturation level in blood flow.
This invention relates to a neural network processing device for analyzing spectral responses to determine oxygen saturation levels in blood flow. The device processes at least two spectral responses, such as those obtained from a light source and a detector, to extract relevant parameters. These parameters are then correlated with oxygen saturation levels in blood flow using a trained neural network. The neural network is configured to interpret the spectral data, which may include intensity, wavelength, or other spectral characteristics, to estimate oxygen saturation accurately. The device may also include additional components, such as a light source and a detector, to capture the spectral responses from a target area, such as tissue or blood vessels. The neural network may be trained using labeled data to improve accuracy in oxygen saturation estimation. This technology is useful in medical applications, such as pulse oximetry, where non-invasive monitoring of blood oxygen levels is required. The device provides a more precise and automated method for determining oxygen saturation compared to traditional techniques.
16. The device of claim 15 , wherein the neural network processing device is further configured to correlate the one or more parameters obtained from the first spectral response and the second spectral response to the oxygen saturation level in blood flow.
This invention relates to a neural network processing device for analyzing spectral responses to determine oxygen saturation levels in blood flow. The device processes spectral data obtained from a first spectral response and a second spectral response, where the first spectral response is captured under a first condition and the second spectral response is captured under a second condition. The neural network processes these spectral responses to extract one or more parameters, such as intensity, wavelength, or other spectral features. The device then correlates these parameters to determine the oxygen saturation level in blood flow. The neural network is trained to recognize patterns in the spectral data that correspond to varying oxygen saturation levels, enabling accurate and non-invasive measurement. The system may be used in medical applications, such as pulse oximetry, where continuous monitoring of blood oxygen levels is required. The neural network enhances the accuracy and reliability of oxygen saturation measurements by leveraging machine learning techniques to analyze complex spectral data. The device may also include additional processing steps, such as noise reduction or signal enhancement, to improve the quality of the spectral data before analysis. The overall system provides a robust solution for monitoring blood oxygen levels in clinical or wearable health monitoring devices.
17. The device of claim 10 , wherein the neural network processing device is further configured to correlate one or more parameters obtained from at least one of the first spectral response and the second spectral response to a glucose level in blood flow.
This invention relates to a neural network processing device for analyzing spectral responses to determine blood glucose levels. The device processes spectral data obtained from a non-invasive measurement system, which includes a light source and a detector. The light source emits light at multiple wavelengths, and the detector captures the resulting spectral responses from tissue, such as the skin. The neural network processes these spectral responses to extract relevant parameters, such as absorption peaks or scattering characteristics, which are then correlated to blood glucose concentrations. The system may use machine learning models trained on reference glucose measurements to improve accuracy. The neural network can also account for variations in tissue properties, environmental factors, or individual differences to enhance reliability. This approach enables continuous, non-invasive glucose monitoring without requiring blood samples, addressing the limitations of traditional invasive methods. The device may integrate with wearable or portable systems for real-time glucose tracking in diabetes management.
18. A device, comprising: a signal processing circuit including a plurality of light emitting diodes and at least one photodetector configured to: obtain photoplethysmography (PPG) signals at a first wavelength from skin tissue of a patient and at a second wavelength from skin tissue of the patient, wherein the first wavelength is in a range of 370 nm-410 nm and wherein the second wavelength is in a visible or an infrared (IR) range; generate PPG input data using the PPG signals obtained from the patient at the first wavelength in the range of 370 nm-410 nm and at the second wavelength in the visible or the IR range; and a neural network processing device configured to: obtain one or more parameters generated from a training set, wherein the training set includes PPG signals obtained from a healthy population at the first wavelength in a range of 370 nm to 410 nm and at the second wavelength in the visible or the IR range and corresponding known glucose levels obtained from the healthy population; receive the PPG input data; and determine a glucose level in blood flow of the patient from the PPG signals obtained from the patient at the first wavelength in the range of 370 nm to 410 nm and at the second wavelength in the visible or the IR range.
The invention relates to a non-invasive glucose monitoring device that uses photoplethysmography (PPG) signals to estimate blood glucose levels. The device addresses the challenge of accurately measuring glucose without invasive blood sampling by leveraging optical sensing and machine learning. The system includes a signal processing circuit with multiple light-emitting diodes (LEDs) and at least one photodetector. The LEDs emit light at two distinct wavelengths: one in the ultraviolet range (370-410 nm) and another in the visible or infrared (IR) range. The photodetector captures PPG signals from the patient's skin tissue at both wavelengths, generating input data for glucose level estimation. A neural network processing device is trained on a dataset of PPG signals from healthy individuals, along with their corresponding known glucose levels. The trained neural network analyzes the patient's PPG signals at both wavelengths to determine their blood glucose concentration. This approach combines optical sensing with machine learning to provide a non-invasive, continuous glucose monitoring solution.
19. The device of claim 18 , wherein the second wavelength includes one of: 660 nm, 880 nm, 940 nm, or 1050 nm.
This invention relates to a medical device for non-invasive blood glucose monitoring using multiple wavelengths of light. The device addresses the challenge of accurately measuring blood glucose levels without requiring invasive procedures like finger-prick tests. The system employs a light source that emits at least two distinct wavelengths of light, where the first wavelength is used for reference measurements and the second wavelength is selected from 660 nm, 880 nm, 940 nm, or 1050 nm to enhance sensitivity to glucose concentration. The device includes a detector that captures light transmitted or reflected from tissue, and a processing unit that analyzes the detected light to determine glucose levels. The second wavelength is chosen to optimize signal-to-noise ratio and minimize interference from other tissue components. The system may also incorporate calibration techniques to account for variations in skin tone, hydration, or other physiological factors. The device is designed for portable, continuous monitoring, providing real-time glucose readings for diabetes management. The use of specific wavelengths improves accuracy compared to single-wavelength systems, addressing limitations in prior art where interference from hemoglobin or other molecules degraded measurement reliability.
20. The device of claim 18 , wherein the neural network processing device is further configured to determine an NO level in the blood flow from the PPG input data.
This invention relates to a medical device for monitoring blood flow and analyzing physiological parameters, specifically focusing on the detection of nitric oxide (NO) levels in blood flow using photoplethysmography (PPG) signals. The device addresses the challenge of non-invasive, real-time monitoring of NO, a key signaling molecule in cardiovascular and metabolic health, which is traditionally difficult to measure accurately without invasive methods. The device includes a neural network processing unit that analyzes PPG input data to extract physiological information. The neural network is trained to interpret PPG waveforms, which represent blood volume changes in the microvascular bed, and derive NO levels from these signals. The system may also incorporate additional sensors or processing modules to enhance accuracy, such as motion artifact correction or signal preprocessing to isolate relevant frequency components. The neural network is configured to process raw or preprocessed PPG data, applying machine learning techniques to correlate specific waveform features with NO concentrations. This allows for continuous, non-invasive monitoring of NO dynamics, which can be used in clinical settings for assessing vascular health, detecting endothelial dysfunction, or monitoring treatment responses. The device may output NO levels in real-time or store data for further analysis, providing a valuable tool for personalized medicine and chronic disease management.
21. The device of claim 18 , wherein the signal processing module is configured to generate the PPG input data from the PPG signals by generating one or more of: a pulse shape, average distance between pulses, variance, instant energy information, or energy variance.
The invention relates to a signal processing device for analyzing photoplethysmogram (PPG) signals, which are commonly used in medical and wearable devices to monitor physiological parameters such as heart rate and blood oxygen levels. The device addresses the challenge of extracting meaningful physiological information from raw PPG signals, which can be noisy and variable due to motion artifacts, sensor placement, and individual differences. The signal processing module within the device processes PPG signals to generate input data for further analysis. Specifically, it derives multiple features from the PPG signals, including pulse shape, average distance between pulses, variance, instant energy information, and energy variance. These features provide a comprehensive representation of the PPG waveform, enabling more accurate and reliable physiological monitoring. By extracting these specific characteristics, the device improves the robustness and accuracy of health assessments in applications such as fitness tracking, medical diagnostics, and continuous patient monitoring. The invention enhances the utility of PPG-based systems by providing detailed signal metrics that can be used for advanced analytics and decision-making.
22. The device of claim 18 , wherein the signal processing module is configured to generate the PPG input data from the PPG signals by generating characteristic features related to a shape of waveforms of the PPG signals.
This invention relates to a medical device for processing photoplethysmographic (PPG) signals to extract physiological information. PPG signals are optical measurements of blood volume changes in tissue, commonly used in pulse oximetry and other health monitoring applications. A key challenge in PPG signal processing is accurately extracting meaningful physiological parameters from noisy or distorted waveforms. The device includes a signal processing module that analyzes PPG signals to generate input data for further analysis. The module extracts characteristic features related to the shape of the PPG waveforms, such as peak-to-peak amplitude, waveform symmetry, and temporal characteristics. These features help distinguish between different physiological states, such as normal vs. abnormal blood flow patterns. The extracted features are then used to derive physiological metrics like heart rate, oxygen saturation, or vascular health indicators. The device may also include additional components, such as a sensor interface for acquiring raw PPG signals and a data output module for transmitting processed data to a display or storage system. The signal processing module may apply filtering, normalization, or other preprocessing steps to enhance signal quality before feature extraction. The extracted features can be used in algorithms for real-time monitoring, diagnostic assessments, or long-term health tracking. This approach improves the reliability and accuracy of PPG-based measurements by focusing on waveform shape analysis rather than raw signal interpretation.
23. The device of claim 18 , wherein the neural network processing device is further configured to: periodically receive an updated learning vector, wherein the updated learning vector is generated from an updated training set, wherein the updated training set includes additional PPG input data obtained using additional PPG signals at the first wavelength in the range of 370 nm-410 nm and at the second wavelength in the visible or IR range and known glucose concentration levels; and reconfigure one or more parameters for generating the glucose level in blood flow of the patient from the PPG input data.
This invention relates to a neural network-based device for non-invasively measuring blood glucose levels using photoplethysmography (PPG) signals. The device addresses the challenge of accurately estimating glucose concentrations from PPG data, which varies due to physiological and environmental factors. The neural network processes PPG signals at two distinct wavelengths—one in the ultraviolet range (370-410 nm) and another in the visible or infrared range—to derive glucose levels. The device includes a PPG sensor module to capture these signals and a neural network processing unit that analyzes the data to output glucose concentration estimates. To maintain accuracy over time, the neural network periodically receives updated learning vectors derived from an expanded training set. This updated training set includes additional PPG signals at the specified wavelengths, along with corresponding known glucose concentration levels. The neural network then adjusts its parameters to improve glucose level predictions based on the new data. This adaptive learning approach ensures the device remains reliable despite variations in patient physiology or external conditions. The system is designed for continuous or intermittent monitoring, providing real-time glucose level feedback without invasive blood sampling.
24. The device of claim 18 , wherein the device is implemented within a user device or is configured to communicate with a user device.
A system for managing user interactions with digital content is disclosed. The system addresses the problem of inefficient content delivery and user engagement by dynamically adjusting content presentation based on user behavior and preferences. The system includes a processing unit that analyzes user interactions with digital content, such as clicks, dwell time, and navigation patterns, to determine user preferences and engagement levels. The system also includes a content delivery module that modifies the presentation of content, such as adjusting layout, prioritizing certain content, or suggesting related content, based on the analysis. The system may be integrated into a user device, such as a smartphone or tablet, or operate as a separate system that communicates with the user device to provide personalized content recommendations. The system may also include a machine learning component that continuously improves content delivery strategies by learning from user feedback and behavior over time. This approach enhances user experience by delivering more relevant and engaging content, while also improving content creators' ability to reach their target audience effectively.
Unknown
January 12, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.